A major challenge in cancer therapy has been the selective and efficient delivery of therapeutic agents to the lesions. Solid tumors such as pancreatic cancer often have poorly-vascularized regions at their cores that are hypoxic and are not readily accessible to the vascular delivery of chemotherapeutic drugs. Hypoxia, however, offers the potential for anaerobic bacterial colonization that can lead to tumor destruction, and one such anaerobic bacterium is Clostridium perfringens (Cp). We have deleted its alpha toxin gene that encodes phospholipase c (Cp/plc-), and showed that the mutant strain can no longer cause gas gangrene in mice. We have also showed that the dormant spores of this deletion mutant, when administered intravenously in mice bearing an orthotopic model of pancreatic cancer, were capable of germination and proliferation in the hypoxic tumor regions with oncopathic effects. However, Cp/plc- does have residual tolerance to oxygen and retains limited capabilities for growth in oxygenated tissues, leading to significant toxicities in animals when administered intravenously at high doses. Superoxide dismutase (sod) is a major oxygen tolerance gene that has also been knocked out in Cp/plc-. The maximum tolerated dose of this knock-out strain (Cp/plc-/sod-) was elevated by one-log over that of Cp/plc- in tumor-bearing mice. To further improve safety, we hypothesize that sequential knock-out of the other major oxygen tolerance genes in the sod knock-out strain will produce an oxygen-intolerant strain (oiCp/plc-) with maximal tumor selectivity and minimal toxicity. However, intratumoral bacteria replication was inhibited by a rapid accumulation of host inflammatory cells that are bactericidal, which limited the extent of tumor response. To enhance oncopathic potency, we hypothesize that intratumoral replication of oiCp/plc- and its antitumor efficacy can be substantially elevated by constructing recombinants that express inflammation suppressive genes from heterologous microbes endowed naturally with such properties. These novel recombinant strains will have maximal tumor selectivity and oncopathic potency in poorly-vascularized tumors. Finally, most solid tumors also contain regions that are relatively well-vascularized and oxygenated, and hence refractory to the anaerobic bacteria treatment. These vascularized and oxygenated regions however, are susceptible to systemically administered chemotherapeutic drugs. Thus we hypothesize that anaerobic bacteria treatment that targets the hypoxic cores of the poorly- vascularized tumors will be complementary to systemic chemotherapy that targets the vascularized regions, leading to substantially enhanced tumor destruction and survival prolongation. Successful conduct of the proposed studies may lead to the development of recombinant Cp spores as a novel class of therapeutic agents that can be administered systemically and safely to patients with advanced pancreatic cancer and other poorly-vascular tumors in the future and which, in combination with chemotherapy, may lead to improved treatment outcome than systemic chemotherapy alone, which is the standard of care at present. PUBLIC HEALTH RELEVANCE: Although pancreatic cancer is only the 10th and 11th leading cause of new cancer cases in U.S. men and women, respectively, it is the 4th leading cause of cancer deaths for both U.S. men and women. 33,370 deaths and 37,170 new cases are estimated for 2007 in the U.S. alone. Currently available treatment modalities include chemotherapy and radiation therapy, which are not particularly effective for pancreatic cancer. Anaerobic bacteria can target the hypoxic cores in tumors that will be complementary to chemotherapy, such that the outcome of combination treatment will be superior to chemotherapy alone.